Upon release from infected cells, immature retroviruses undergo a maturation process in which Gag is cleaved by the viral protease into MA, CA, and NC, triggering large morphological changes and producing infectious virions. Within the mature virion, MA remains associated with the viral envelope, while CA assembles as a fullerene cone with ~250 CA hexamers and 12 CA pentamers, which encloses the RNA genome complexed with NC. In the last funding cycle we determined a 9-E resolution cryoEM density map of full-length HIV-1 CA by electron crystallography of 2D crystals. Docking of high-resolution structures of the N- terminal domain (NTD) and C-terminal domain (CTD) yielded a molecular model that guided the insertion of disulfide bonds that stabilized the NTD hexamer. Further mutagenesis destabilized the CTD dimers that link adjacent hexamers, thereby enabling solubilization and 3D crystallization. The resulting atomic-resolution X-ray structures revealed that the CA hexamer is composed of a relatively rigid inner ring of NTD subunits, surrounded by a mobile belt of CTD subunits. Mobility of the CTD belt is likely to be an underlying mechanism for generating the continuously curved capsid lattice in the fullerene cone. The same disulfide strategy was then used to generate stable CA pentamers, and we are completing the first high-resolution X-ray structure. For the next funding cycle we will pursue 3 specific aims: (1) We will devote 40% effort to continue our structural studies of the mature capsid lattice. In addition to completing X-ray structures of the pentamer, we will determine subnanometer cryoEM reconstructions of CA tubes with variable diameters. With high-resolution structures of the hexamer and pentamer, and guided by the hexamer interactions in the CA tubes, we will use computational methods to build an atomic model for the conical capsid. (2) We will devote 40% effort to structural studies of the immature Gag lattice. We have generated Gag mutants that display helical diffraction and serve as an in vitro mimic of the immature lattice. By analogy with our studies of the mature lattice, cryoEM and molecular docking will yield a model that will guide the engineering of soluble Gag hexamers for cryoEM and X-ray crystallographic studies. (3) We will devote 20% effort to explore the structures of possible maturation intermediates during assembly. We have generated multiple 2D crystal forms of a Gag-like construct of murine leukemia virus that display subnanometer diffraction, and our working assumption is that analysis of these varying crystal forms may reveal structural rearrangements that occur during maturation. We are hopeful that our structural studies will continue to provide insight into principles of retrovirus assembly that will be important for the design of new therapeutic strategies.